Pay for what performance?

Posted by Marc Hodak on November 11, 2013 under Executive compensation, Governance, Invisible trade-offs, Pay for performance | Read the First Comment

Pay for performance seems like such a simple idea, and easy to accept as a basis for judging executive compensation.  So why does it continue to create such discussion and controversy?  Well, consider the following grid:

The key distinction is managerial performance versus company performance.  An easy way to understand this distinction is to consider a gold mining company when the gold price has dropped significantly, but our company’s profits and stock price have dropped less than half of anyone else’s in our sector due to extraordinary management.  It’s easy to see in such an example that our management has done great, but our shareholders have done poorly.  Should such managers get a bonus?

When management and shareholder performance are strong, as in the upper right quadrant, the answer is obvious.  When management and shareholder performance are weak, as in the lower left quadrant, the answer is obvious.

But what do we do when our gold company finds itself in the upper left quadrant?  If we pay a bonus for this situation, we are open to the accusation of pay without performance by our investors.  Our investors might bother to look at relative performance, in which case they might forgive bonus payments up to a point.  But there is no way outside investors can gauge what the board can, i.e., that our managers actually did a great job given their situation, and that denying them a bonus may entail a significant risk of losing them to other firms that promise to compensate them for being great managers.

Paying a bonus for lower right quadrant performance is equally problematic.  Most shareholders will let you get away with it because they are feeling flush.  But those that don’t are on firmer ground in saying it would be pay without performance, that management was simply in the right place at the right time.  In this situation, the downside to not paying a bonus is a little more subtle.  If we deny managers a bonus for poor relative performance in the face of good absolute performance, then we MUST be willing to pay them bonuses for good relative performance even when the company suffers poor absolute performance.  In other words, boards justifiably refusing to pay bonuses when they are in the lower right quadrant will eventually they find themselves in the upper left quadrant having to pay bonuses, or risk almost certain loss of their best managers.  And we already highlighted the difficulty in adhering to a policy of consistently paying for relative, as opposed to absolute performance.

True to their pragmatic form, many boards resolve this dilemma by paying for both absolute and relative performance.  This makes the plans more complicated, and does not completely eliminate at least some criticism of pay without performance, but it at least attempts a workable compromise.  Fortunately, ISS (pdf) and Glass-Lewis provide a least some cover for pay for relative performance, but that only gets you so far.

What would you do?